[Talk-ca] Licence de données ouvertes, Montréal
penorman at mac.com
Sun Mar 3 22:07:26 UTC 2013
> From: Harald Kliems [mailto:kliems at gmail.com]
> Sent: Sunday, March 03, 2013 7:28 AM
> Subject: Re: [Talk-ca] Licence de données ouvertes, Montréal
> On Sat, Mar 2, 2013 at 3:58 PM, Paul Norman <penorman at mac.com> wrote:
> > Lastly, cadastral data is probably the least exciting type of data for
> > Other data like roads, addresses and even buildings is more useful.
> Oh, I thought cadastral data would include building outlines? At least
> that's the case with the somewhat controversial French cadastre imports.
> Did I get excited prematurely?
The exact definition varies, but the generally accepted definition in North
America is the zoning, tax or property lot information. It does not include
buildings, roads, waterways, sewer infrastructure, etc.
It does not generally contain addresses because there is not a one to one
relationship between cadastral areas and addresses.
I was talking with several people last week about an integrated cadastral
fabric for BC.
Cadastral information is actually fairly useful as open data, but most
people who would want to use it would be forced to get it from the city
anyways. There can be some information useful to OSM in it, but generally
most of the information is not relevant to us. I tried extracting useful
information from Surrey's cadastral layer and found it a lot of work.
I did a presentation on open data and OSM to an audience with many open data
publishers and I ranked the most useful data as orthophotos, addresses and
roads. Orthos because good ones are hard or expensive to get and cities
often have excellent ones. Addresses because they're useful for geocoding,
annoying to collect manually, and well suited to conflating with existing
data. Roads because the road network is generally considered to be one of
the most important parts of OSM, so even if the roads are only marginally
better than CanVec they can still be useful.
More information about the Talk-ca